The short answer
They work. The median advertiser account pays $14.83 for a landing page click, $140.49 for a completed lead form submission and $350 for a demo request. Those are workable numbers for a business selling a $20,000 contract. They are terrible numbers for a business selling a $200 one.
So the honest answer is that LinkedIn Ads work when your deal size can carry the cost, and the single best predictor of whether they will work for you is your average contract value, not your creative.
The more useful finding is what separates the accounts where it works from the accounts where it does not. It is almost never the ads.
Every figure here comes from our own panel: 1,000+ advertiser accounts, 22,942 campaigns and $79 million of measured LinkedIn spend across 1.34 billion impressions in the twelve months to 7 September 2026. Medians across advertiser accounts, never pooled totals. LinkedIn Audience Network delivery excluded.
The 121x spread, and what explains it
We took 119 advertiser accounts where we could measure the entire path, impression to booked meeting, and compared the cheapest to the most expensive.
The gap was 121 times.
Then we decomposed it. Traffic price, meaning CPM and cost per click, explained 1.5 times of that gap. Landing page and form conversion explained 72 times.
Read that as a sentence: the thing every agency optimises accounts for roughly one fiftieth of the variance in what actually matters. Bidding, audience tuning and creative testing move the 1.5. What happens after the click moves the 72.
This is why two companies can run near-identical campaigns and get results an order of magnitude apart, and why the one with worse ads sometimes wins. LinkedIn Ads do not fail in the auction. They fail on the landing page, in the form, and in the hour after the lead arrives.
Most accounts are missing a funnel layer
Only 35.7% of advertiser accounts run all three funnel layers: cold reach, retargeting, and a conversion mechanism. Nearly two thirds are missing one.
The missing layer is almost always the middle. Companies run cold awareness campaigns and a book-a-demo page, with nothing in between, then conclude the channel does not convert.
It does not convert because nobody books a demo with a company they met ninety seconds ago. LinkedIn is a demand creation channel being asked to perform as a demand capture channel, and the gap between those two jobs is the retargeting layer that two thirds of accounts never build.
If you tried LinkedIn Ads, ran cold traffic to a demo page, and stopped, you did not test LinkedIn Ads. You tested one third of them.
88% of lead form opens are abandoned
Lead gen forms open at $18.21 and complete at $140.49. The completion rate is 11.6%.
So 88 out of every 100 people who tapped your form, who were interested enough to start, left without finishing. That is a 7.7x difference in cost sitting inside a single campaign, and it is the cheapest problem to fix on this entire list.
The usual cause is field count. Every field added because somebody in sales asked for it gets paid for at the completion rate, not the open rate. A phone number field is not free. It costs you a measurable share of the 11.6%.
It is also the most common reporting error we find. LinkedIn exposes both numbers, and agencies quote the one that flatters. If your reported cost per lead looks suspiciously good, check whether you are being shown form opens or form submissions.
Yes, it is expensive. That is a different question
LinkedIn is expensive against Meta and Google on every raw metric and it always will be. The median CPM is $57.38. In the United States it is $92.01.
But cost per thousand impressions is not a business metric. The question is cost per closed deal, and on a $20,000 contract a $350 demo request is an excellent trade. On a $2,000 contract it is not a trade at all.
The market you buy in moves this more than most people realise. The United States costs 2.3 times the United Kingdom per landing page click, $46.98 against $20.69, and 2.1 times per lead. A UK company told to lead with US spend is paying double for the same outcome, and that decision is usually made on ambition rather than arithmetic.
When LinkedIn Ads genuinely do not work
There are real cases, and it is worth naming them rather than pretending the channel suits everyone.
If your average contract value is under roughly $5,000, the maths rarely closes. If your buyer is not identifiable by job title, company size or industry, LinkedIn’s targeting advantage disappears and you are paying a premium for nothing. If you need results inside thirty days, a demand creation channel is the wrong tool and paid search is the right one. And if nobody owns what happens after the click, you will reproduce the wrong end of that 121x spread no matter what you spend.
Everything else is a setup problem, and setup problems are fixable.
Related questions
Are LinkedIn Ads worth it?
For B2B companies with contract values above roughly $20,000, yes: the median cost per demo request is $350. Below about $5,000 of contract value the arithmetic rarely works, regardless of how well the campaigns are run.
Why are my LinkedIn Ads not converting?
Most likely the funnel, not the ads. Only 35.7% of advertiser accounts run all three layers, and the missing one is usually retargeting. After that, check your lead form: 88% of opens are abandoned across the panel.
How long before LinkedIn Ads work?
Treat ninety days as the minimum honest read. LinkedIn creates demand rather than capturing it, so the first thirty days build the retargeting audiences that the next sixty convert. Judging it on month-one cost per lead measures the wrong thing.
Are LinkedIn Ads better than Google Ads?
They do different jobs. Google captures demand that already exists; LinkedIn creates it among people who are not searching yet. Most B2B programmes that work run both, with LinkedIn feeding the searches that Google then captures.